The SRTR reports regularly about center-level differences in graft failure after deceased donor kidney transplantation (DDKT). While these models are adjusted, they provide only information about outcomes on average, across all donor types. However, the most heterogeneity in practice patterns exists at the extremes of donor risk. The goal of this study was to evaluate the degree of outcome variation for high-KDPI DDKT above and beyond existing center-level differences.

Methods: Using SRTR data on 81,446 adult, first-time, kidney-only DDKT recipients 1/1/2005-12/31/2014 from 219 transplant centers, we estimated the overall association between KDPI>85 and graft loss, as well as a unique hazard ratio of KDPI>85 for each center, using multilevel Poisson regression (random intercept + random slope for KDPI>85). To test whether center-level differences in risk for KDPI>85 were statistically significant above and beyond overall center-level differences in graft failure, we compared this model to a random-intercept-only model using a likelihood ratio test.

Results: Overall, DDKT from a donor with KDPI>85 was associated with 60% increased risk of graft loss (HR =_1.541.60_1.67). The association between KDPI>85 and graft loss varied slightly by center from HR of 1.38 to 1.82 (figure) ; however, these center-level differences were not statistically significant (p>0.9).Conclusions: There is no evidence that centers differ in outcomes following transplantation with high-KDPI kidneys above and beyond differences in performance already captured by the SRTR. As such, regulatory report cards would be expected to look exactly the same with or without inclusion of these high-risk organs.

CITATION INFORMATION: Wu L, Massie A, Luo X, Bowring M, Segev D. Center-Level Differences in Outcomes Following High-KDPI Kidney Transplants. Am J Transplant. 2017;17 (suppl 3).

« Back to 2017 American Transplant Congress